Theory in ecology predicts that species with a large number of individuals will be larger targets for diseases. Parker et al. (2015) tested the idea in grassland plants. They measured the abundance of each of 43 species of plants (calculated as percent cover) and used this to predict the mean percentage of tissue on the plants that showed signs of disease. A scatter plot of their data is shown below. The data are available online at whitlockschluter3e.zoology.ubc.ca

a. Comment on whether these data fit the assumptions of linear regression. If not, what solutions would you suggest?

b. Use the log transformation to transform these data. Then test the null hypothesis of no relationship between the percentage of tissue showing disease and plant abundance. Does abundance predict the mean % disease of a grassland plant species? Assume that the species data points are independent.

c. What is the 95% confidence interval for the slope of the relationship between log abundance and log % disease?

Transcribed Image Text:

Mean % disease 30 20 10 0 0.0 0.2 0.4 Abundance 0.6